RisorseArtificiali

Compare the current agentic readiness state against a previous v2 assessment. Shows per-dimension deltas (📈/📉/➡️), overall delta, and Portable/Target layer deltas, citing the files that drove each change. v2-schema only: if no comparable v2 baseline exists, runs a fresh scan instead of diffing. Use after making changes to measure progress.

Remediate a project's agentic-readiness gaps by generating missing, project-specific files: AGENTS.md (primary, vendor-neutral) + target bridges, .env.example, .gitignore secret coverage, docs/agent-execution.md, CI + pre-commit baselines, Dependabot/Renovate, specs/ADR/issue templates, Makefile targets, and a generated repo index. Reads prior scores, prioritizes by impact, acts only on skill/partial-fixable sub-criteria, surfaces manual ones, and shows a before/after delta. Use after /agent-ready scan to close identified gaps.

Scaffold a minimal, opinionated, agent-ready baseline for a new (or near-empty) project. Generates a portable-first AGENTS.md (<200 lines) with target bridges, .env.example, .gitignore secret coverage, a documented agent-execution/sandbox policy (LINCE among options), CI + pre-commit baselines, and an optional spec template — then runs a quick v2 scan to show the starting score. Use for greenfield projects; defers to /agent-ready fix on a populated existing repo.

Render the v2 agentic readiness assessment as a single layered Markdown report in .agent-ready/. Includes an executive summary with ASCII bars, Portable-vs-Target layer analysis, per-dimension detail with explained findings (why / consequence / how-to-fix / effort) for every sub-criterion below 100, and a remediation roadmap. Generates a README badge and supports --format html. Use after a scan to produce a shareable readiness report.

Full diagnostic scan of a project's agentic coding readiness. Evaluates 7 evidence-weighted dimensions (agent instructions, navigability & code intelligence, testing & feedback, CI/CD & governance, agent tooling, security & sandbox, spec-driven workflow & docs) with portable-vs-target layering, an optional --agents target list, and helper-script signals, producing a 0-100 score plus explained findings. Use when you need to evaluate how AI-ready a codebase is.

Assess and improve a project's readiness for agentic coding. Scores 7 evidence-based dimensions (agent instructions, navigability & code intelligence, testing & feedback, CI/CD & governance, agent tooling, security & sandbox, spec-driven workflow) on a 0-100 scale with portable-vs-target layering and explained, fixable findings. Use when evaluating how well a codebase supports AI-assisted development, scaffolding a new agent-friendly project, or when the user mentions 'agent ready', 'agentic readiness', 'AGENTS.md', or 'AI-ready project'.

Suggerisce data-driven quale video mettere in YouTube end screen (1 video, layout subscribe+video) e quali 5 video del canale linkare con YT cards a timestamp specifici del video corrente, per il podcast "Risorse Artificiali". Match semantico LIGHT tra metadata episodio corrente (titolo + description + chapters) e metadata episodi passati nel repo Jekyll (titolo + description + chapters H2), arricchito con metadata yt-dlp (views, durata) per tiebreaking. NON legge transcript completi (troppo costoso in token). Output: capitolo "End screen + YT Cards" da appendere al promo file consolidato di podcast-promo. Attiva quando l'utente vuole configurare end screen o cards di un nuovo episodio YT, suggerire link interni del canale, o ottimizzare la session duration via cross-linking. Invocazione manuale dopo podcast-promo v4.2.

Skill master per la pubblicazione di un nuovo episodio del podcast "Risorse Artificiali". Un'unica invocazione produce titolo high-CTR (pattern numerato o intervista), frasi in sovraimpressione verbatim per il video (7 brevi + 3 lunghe con timestamp), brief thumbnail con prompt image pronto per ChatGPT Image 2 (con face reference photo del soggetto), chapters YouTube, descrizioni YouTube/Spotify + tag SEO (review unificata), script Shorts + spec Spotify Clip (review unificata), post LinkedIn, sezione newsletter codiceartificiale, e per le interviste un Guest Launch Kit completo. In output scrive DUE file: un promo file consolidato (`podcast-promo/episodes/{date}-{slug}_promo.md`) con tutti i contenuti pronti copia-incolla, e il post Jekyll (`_posts/{date}-{slug}.md`) con frontmatter v3.0 e trascrizione verbatim. Attiva quando l'utente vuole pubblicare un nuovo episodio, preparare il drop di una puntata, menziona Riverside, o materiali episodio.

Generate professional, comprehensive interview briefings following a structured 5-step process. Use this skill when users want to prepare for interviews by creating detailed briefings that include biography research, thematic blocks, strategic questions, and conducting notes. Ideal for podcast hosts, journalists, researchers, and anyone conducting 30-90 minute professional interviews who needs systematic preparation.

Genera il pacchetto completo per il rilancio retroattivo di un'intervista già pubblicata del podcast "Risorse Artificiali". Produce nuovo titolo (pattern [hook] | [guest]), brief per nuova thumbnail, nuova descrizione YouTube e Spotify, reflection post LinkedIn dell'host con angle callback "n mesi dopo", e un Guest Re-Launch Kit completo per l'ospite (post LinkedIn e X in prima persona, blocco "aged well / aged poorly", email accompagnatoria). Orchestra le skill podcast-promo, thumbnail-gen e podcast-transcript retrofit. Attiva quando l'utente vuole rilanciare un'intervista esistente, fare un "throwback", aggiornare metadati di un episodio vecchio, o ha bisogno di un reflection post per una conversazione passata.

Aggiorna (modalità B, retrofit) post Jekyll esistenti del podcast "Risorse Artificiali" allo schema Frontmatter Convention v3.0 (layout episode, episode_number, spotify_episode_id, guest_*, resources, duration ISO 8601). Use case principale: aggiungere apple_episode_url dopo auto-publish Apple, correzioni retroattive, migrazione post storici v2.0 → v3.0 schema, support al retrofit lanciato da `interview-relaunch`. La modalità A (creazione nuovo post ex novo) è LEGACY da v3.2: per nuovi episodi usa `podcast-promo` v4.0 che scrive direttamente il Jekyll post unificando promo + post.

Genera prompt pronti per modelli di image generation moderni (Gemini 3 Pro, ChatGPT Image 2, Ideogram, Flux 1.1 Pro, Midjourney v7) per creare thumbnail YouTube 16:9 (1280×720) del podcast Risorse Artificiali con testo integrato nel prompt, template differenziati numerato vs intervista, regole anti-necrologio. Use case: iterazioni creative aggiuntive su una thumbnail non convincente (3 varianti alternative), generazione thumbnail per rilanci (invocata da `interview-relaunch` Passaggio 3), produzione visual non-episodio (social graphics, cover articoli indipendenti). Per nuovi episodi del podcast usa direttamente `podcast-promo` v4.0 che include brief thumbnail + 3 prompt pronti al Passaggio 13 (non serve invocare separatamente).

Run and verify code in a disposable, isolated lab VM, and bisect regressions. Use when asked to test a fix in a clean machine, drive the lince quickstart wizard, check install.sh idempotency, reproduce a bug in isolation, or find which commit broke something. Authors a recipe TOML, runs it via the lince-lab CLI (broker over a unix socket), and reads the result artifacts — never runs limactl or touches the host filesystem directly.

Configure LINCE sandbox and dashboard settings in natural language. Use when asked to set up providers, change security levels, modify agent settings, configure API keys, adjust sandbox behavior, or any "how do I configure" question about LINCE. Drives the Config v2 policy layer (~/.config/lince/lince.toml) through the lince-config CLI — apply/set/validate only, never direct TOML edits.

Add support for a new AI coding agent (or CLI tool) to lince-dashboard and agent-sandbox. The agent provides its own requirements (binary, config dirs, API keys, sandbox behavior) and this skill generates correct TOML configuration. Use when adding a new agent type, setting up multi-agent support, or when asked to add agent, new agent type, support cursor, support bob, support cli, register agent, setup agent, configure agent for dashboard.